No way!? Is that ‘Possible’?


There is an argument for the existence of God called ‘The Modal Ontological Argument for the existence of God’ (MOAG). A major part of MOAG is the idea that something is ‘possible’. The “something” is a ‘maximally great being’, or something similar depending on which version you read. The vagueness of the “something” ― and the difficulty in defining it ― is a problem with MOAG, but the defense of whether it is ‘possible’ is worse.

There are not many ways to establish if something is possible: you can either induce the possibility of some phenomenon by demonstrating that the general category of phenomena has already occurred; establish that the component parts of a phenomena are individually possible; or show that the something is a direct extrapolation of reliable models of reality we already have.

In addition, there are ways of trying to establish possibility that are simply inadequate. The definition of ‘Possible’ that pops up a lot listening to people defending various MOAGs is ‘not demonstrated to be impossible’ or ― to do away with simply defining something as ‘not its negation’ ― ‘not ruled out by logical laws’.

So, let’s go through these: what do the three examples of demonstrating possibility look like?; and why is the ‘not demonstrated to be impossible’ definition of ‘possible’ inadequate?

General categories

If you think that something is possible because similar things have happened before, then you are making what I am calling a ‘general category’ demonstration of the possibility of something.

It is possible that a meteorite could hit Earth. We know they have, so we know they can. In 2013, one hit Russia. There are problems with this, like that it doesn’t tell us anything about which meteors will become meteorites and strike Earth; for that, we need ‘extrapolated possibility’, using predictions from models.

It also has the ‘Induction Problem’ at its heart: trying to decide what speculative event is like what. Defining categories is often done with an intuitive touch, and so knowing what is demonstrated to be possible by some nebulous category is not actually easy. For example, if your camera can take a high resolution picture of a golf ball at 60 meters, does that mean it is possible for my camera to do the same?

With a question like this, the ‘general category’ relies on some explanatory models. Perhaps my camera doesn’t belong to the relevant category: it lacks an optical zoom or sufficient resolution. The induction problem ― deciding how to define the relevant categories ― relies on some definition of the category that partially explains why something is possible.

Going back to the meteorites, it is not possible for a meteorite that has already struck Earth to strike it again; the relevant category is of meteors currently moving at pseudo-random in space, and an already landed meteorite is excluded from that. That’s an example of the category being defined, at least in part, by the explanation.

Possible components

Is it possible to name a child ‘Herblederble-M’furblekerble’? Such a phenomenon doesn’t really belong to any category, as children’s names tend to be either existing names or real words: “Dillon”, “Apple” and so on. Naming appears to be limited to traditional examples, or things that somehow sound like names. For example, the first time I ever heard the name ‘Rowanna’, I knew it was a name.

‘Herblederble-M’furblekerble’ violates this category. That doesn’t mean it is impossible, all it means is that it is not demonstrated to be possible by the ‘general categories’ approach.

However, naming a child is built of smaller steps, and if the smaller steps are all possible, then the whole process is possible. Naming a child is a case of there being a child currently without a name that you have the right to name. This happens if it’s your child, through birth or adoption. An easy step.

Names are simply larger samples of language built from smaller units of language. The fact I can type out ‘Herblederble-M’furblekerble’ with a standard keyboard, and then pronounce it using phonetic rules of English, demonstrates that it is a possible (non-traditional) name.

To be a name, as opposed to just a nickname, it has to be written on their birth certificate. There are no naming laws in the UK, but the Registering Office has discretionary powers to not accept names that are misleading (e.g. “Doctor”), contain numerals, are impossible to pronounce or contain obscenities. ‘Herblederble-M’furblekerble’ passes the test as all the component parts are possible.

Extrapolated possibility

We knew about black holes before we saw any evidence for them. Before we saw gravitational lensing, gravity waves from the collision of black holes or Hawking radiation, we knew black holes exist. And that is because they were an extrapolation from an effective model of reality: Relativity.

The basic idea of relativity relevant here is that heavier and denser things have a higher gravitational effect. The question, then, was one of what happens if you take any finite space and increase the mass inside it. The answer is that the force of gravity becomes so great that it would collapse all of the mass into a smaller and smaller point, collapsing the structure of any material or even atom. The gravity of that point creates an infinite ‘well’ in spacetime and an horizon from which light couldn’t escape. It is a mathematical prediction of an effective theory; black holes are possible, given the right conditions. Could those conditions exist?

In the life-cycle of a star, its size is the result of outward pressure created by heat and inward pressure created by gravity, because of the star’s mass. As a star runs out of fuel, it cools. If you cool the star and leave gravity at a constant rate, it shrinks and so becomes more dense (same mass, less space). This means gravity is not actually constant, but increases, making the star more dense again, increasing gravity again. This creates a very high pressure, which can create heat and mean the star finds a new steady size, but for very large stars with a lot of mass, gravity simply overpowers the other forces and crushes a star into a black hole; the conditions are possible too.

So, if these are ways of demonstrating possibility, what doesn’t work?

It’s not impossible

From a philosophical point of view, defining ‘possibility’ as ‘not impossibility’ shifts the burden of proof. When someone makes the proposition that something is possible, the definition used puts the onus on the other person to demonstrate impossibility in order to argue. Lay-philosophers, debaters and the casually interested will notice that the onus is placed somewhere other than on the claim-maker, and that is wrong.

We recognise, in most situations, that if I claim something then it is on me to demonstrate it (if anyone cares). Modest claims are sometimes accepted simply by being uttered; the trust we put into strangers is sufficient for us to believe them if they say they have a pet dog. But, if we doubted them, we would be suspicious of anyone who said ‘demonstrate that I don’t have a dog’. And when we’re talking about what is possible, if the discussion is worth having, we are not talking about modest claims where someone’s say-so is all we need.

But the problems with this definition of ‘possible’ are practical as well as philosophical. Is it possible to travel backwards in time? Is it impossible to do this? I don’t know. From a rhetorical point of view, this is important nuance. You can imagine a coarser dialogue, ignoring nuance, between two people (A & B):

A: Is it possible to travel backwards in time?
B: I don’t know
A: Aha! So it must be impossible

And yet the opposite conclusion is reached if you just start with the other question.

You could even look at a real conversation that happened on The Atheist Experience between Matt Dillahunty (D) and a caller (C):

C: It is possible that a God exists.

D: How did you determine that?

C: Well, it’s not impossible.

D: How did you determine that?

C: Can you prove it is impossible?

D: No, no. That’s not how it works…

And that’s not how it works. Notice that the caller implicitly defines ‘Possible’ as ‘Not impossible’ and places the burden of proof on Dillahunty: the caller will accept that a God is possible, until such a time that Mat proves it impossible. But the onus should be on the caller.

What this means to illuminate is that the definition is inadequate for the job. By defining possibility as not its negation you implicitly accept all things as possible until the claim is negated, which shifts the burden of proof; but you also obscure an important distinction: that if something cannot be demonstrated to be impossible, this may be an artefact of ignorance (globally or locally) instead of evidence for the contrary.

Not ruled out by the laws of logic

Sometimes the definition of ‘possible’ will be limited to ‘logically possible’, and, in turn, will include everything you can imagine so long as it not ruled out by logic.

There are many problems with this. For one, you may be justified in evaluating the claim ‘that which is excluded by the laws of physics is impossible’ as a logical claim, but it is not actually within logic itself. Physically impossible things, like something exerting less gravitational force as its density and mass increase, are actually possible under the definition of ‘not ruled out by the laws of logic’; the claim does not violate ‘contradiction’, ‘excluded middle’ or ‘identity’.

Put simply, logical possibility simply doesn’t relate to physical possibility.

The Piltdown Man ― originally, and immediately controversially, touted as an early hominid and part of human evolution ― was discovered to be a hoax, not because it was ‘logically impossible’ for the anatomy of the Piltdown Man to be real, but because it was ‘evolutionarily impossible’. (A fun irony, given how much Creationists liked to use Piltdown Man as an example of corruption and therefore falsity in the science of evolution.)

Logical possibilities are like a first test. Something can be logically possible, but still actually impossible: like an inverse relationship between mass and gravity, or the Piltdown Man. It is a larger subset than ‘possibility’, the level 1 test to see if a discussion is worth pursuing a little further.

It’s another fun irony that the Christian God does appear to violate logical laws: the father, the son and the Holy Ghost? The immaterial God that ‘comes down’ (a material act) to Babel?


All of this is to demonstrate a simple point. It is not obstinate defiance on the part of sceptics to reject the premise of MOAG that claims something possibly exists. It’s not a sudden and convenient inability to distinguish the possible from the impossible (or at least have a framework in place). Quite the opposite: it is the acknowledging that such a non-arbitrary framework does exist, and the premise is never defended in a compliant way.

By insisting on defining “possible” with a backwards burden of proof (“not impossible”) or in the broadest possible terms (“logically possible”) the debater is implying they have some awareness of the fatal weakness of the premise.

49 thoughts on “No way!? Is that ‘Possible’?”

  1. Just to pick that nit, I claim that it is NOT impossible that a meteorite which has struck the Earth can strike the Earth again. In fact, it is trivial. We have meteorites which have been found. Take one out over a patch of ground, drop it, and it will hit the Earth again.

    Yeah, I know trivial and not what you were intending to illustrate. But I claim that although not at all likely, given the resources, I could indeed cause a meteorite to strike the Earth again. I would “just” need to get a meteorite found on Earth, take it up in a spaceship, and direct it towards the Earth again.

    What this boils down to is pretty much anything is impossible if one or more necessary elements are not available, and anything is possible if all necessary elements are present. And all we can say about the presence of critical elements is “they do not appear to be present at this time, in this place and under these conditions”.

    1. A great point.

      I should have gone with a burned matchstick reassembling itself.

      We can go further with these — ‘not possible, yet’, where we have principles in place but lack the technology or infrastructure. (As an example, just imagine what we do basically daily on the internet now, 100 years ago.) But that wasn’t strictly relevant to the MOAG.

  2. It is possible that you have overlooked a couple of the other important concepts when considering the existence of God. The discussion moves toward “plausible” and “probable” after establishing which things are not “impossible”.

      1. Let’s not pull any punches.
        Let’s say the something is God.

        Given that the possibility of God cannot be established, how do you step right out into the world of plausibility?

    1. Just noticed, I nearly let you get away with say “after establishing which things are not “impossible”” in your first comment, and then denying that you think “something should be considered possible, until it is demonstrated to be impossible” in your second comment (in this thread).

      Could you clarify that for me? Why do you move on to questions “plausible” and “probable” after failing to establish “not impossible”? Surely you have to demonstrate possibility first.

      1. Is it impossible for Santa Claus to exist?
        That depends on what you mean by “Santa Claus”.
        If Santa is a magical being that cruises around in a flying sleigh dropping presents, then yes, it is impossible for him to exist.

        If Santa is a man with a white beard in a red suit posing for pictures at the mall, then no, his existence is not impossible. We can now discuss the plausibility of this Santa.

        It is important to define what you mean by “God” before discussing the possibility of its existence.

      2. Alright. Well, I’ve got no stake in defining God; I’m not religious.
        I follow the religious person’s definition of a God in these discussion.
        So, I give the floor to you. Is it possible for your definition of a God to exist?

      3. Identifying as “non-religious” doesn’t remove the responsibility of defining what you mean by God. You must have a definition of God. It’s central to your non-religious status. You can’t be agnostic without defining God.

      4. Not knowing what people mean by “family” is a pretty good reason for not believing in it.

        Not knowing what people mean by “music” is a pretty good reason for not believing in it.

        This is precisely why atheists rarely move the conversation beyond “Possible vs. Impossible”.

      5. Yeah, Allallt, it’s now your job to correctly define Branyan’s God to his satisfaction first before his ability to defend justifying his religious belief is called upon. How do you think you’ll do? Golly gee whiz, that a real puzzler, that one is.

        This is a classical avoidance tactic used by Branyan all the time while pretending this is a problem fundamental to atheism, a believer who wishes you to wave goodbye to any relative and meaningful discussion about probability and plausibility of his religious beliefs while substituting this notion that the problem only belongs to those who might question the validity of his claims.

      6. Yet he will continue to insist that This is precisely why atheists rarely move the conversation beyond “Possible vs. Impossible”! Yup, no cognitive dissonance here. Move along.

      7. I think it’s important to note that the progress in conversation in no way requires him to define his terms or address questions, because conversation was already ceased from progressing by asking him questions and to define his terms.

        It’s like chapter out of a Joseph Heller farce.

      8. If you tell me I’m in your family, I’m going to have to see your definition of family before I believe you.
        I am okay operating in blurred line of whether a 9th cousin is really worth calling ‘family’. It’s not ideal — but language works because approximate meanings are implicit.
        If you tell me the sound of glass and nails falling down stone stairs is music, I’m going to need to see your definition of music before I believe you.
        That is despite the blurred lines caused by whale “song” and ‘Panic at the Disco’. Beacuse approximate definitions are implicit.

        If you tell me it is possible a God exists, I’m going to need to see your definition of a God.
        Because, actually, the unbelievable variations of deism, theism, polytheism, pantheism and panentheism mean there is no implicit approximate agreed meaning.

        So, to lay out what I suspect the agenda of this conversation should be:
        The aim I inferred you have is to demonstrate God is plausible, without demonstrating It is possible.
        To get here, you have to define God first.

        Are you interested? Or are you here for the merry-go-round? Because, I’m not interested in the merry-go-round, I’m interested in the discussion relevant to this post.

      9. So we agree that definition matters.

        No matter how I define God, you can say “Impossible”. I will concede that point and we can get off the merry-go-round.

      10. Not necessarily. You could define God as ‘a subjective sense of purpose’ and I’d have a whole other set of questions that would belong to a different thread altogether.

        Are you going to try to present a definition?

      11. Cool.
        Does being ‘personal’ or a ‘force with agency’ matter, at all? Or could it be a blind force, like physics?

        Is this a case of ‘God is the explanation for the existence of the universe’, no matter what the content of that explanation is, or do you have specific details to that explanation in mind?

        Given the possibility of a multiverse (‘extrapolated possibility’ from effective theories in cosmology and quantum mechanics — particularly Everett interpretations of QM), what do you mean by ‘universe’? This bubble of universe, or all ensembles, however they may be?

      12. This is more like an interrogation than a dialogue. “ANSWER THE QUESTIONS, THEIST! EXPLAIN YOURSELF!” (BTW, that’s why I call Tildeb “Dear Leader”)

        Of course I will answer your questions. I’ll keep talking until I say something you disagree with. You’ve asked several questions so there’s a good chance one of my answers will be objectionable.

        Does being ‘personal’ or a ‘force with agency’ matter, at all? Or could it be a blind force, like physics?
        Agency is important. The laws of physics do not operate with intention. The existence of intelligence and consciousness needs an explanation. Intelligence does not rise from chaos. Mindless forces do not create order.

        Is this a case of ‘God is the explanation for the existence of the universe’, no matter what the content of that explanation is, or do you have specific details to that explanation in mind?
        I don’t understand this question.
        If you’re asking me to explain how God exists, I can’t. But I don’t believe God is a dude in the sky chucking lightning bolts. Whatever attributes are necessary to bring the universe into existence, those I ascribe to God.

        Given the possibility of a multiverse (‘extrapolated possibility’ from effective theories in cosmology and quantum mechanics — particularly Everett interpretations of QM), what do you mean by ‘universe’? This bubble of universe, or all ensembles, however they may be?
        I’m calling physical reality “the universe”.

        If you’re suggesting the “ensembles of universes” outside our own could have agency, intelligence and timeless existence, I would call that “God”.

      13. Okay, so the definition of God has expanded slightly: it is now the explanation, by intention, of the existence of the universe, intelligence and consciousness.

        I’m happy to accept that as the definition of God when talking with you. (Do let me know if I am misrepresenting you.)

        A couple of caveats, of course:
        (1) this is not me accepting that the universe, intelligence or consciousness requires an explanation with intention — just that if that intentional explanation can be demonstrated, the personal force with intention that successfully created it all is called “God”.
        (2) We have not yet ruled out unintentional explanations of all the products discussed.
        (3) There is no specific theology applied.
        (4) There is no reason given to discount some form of pantheism, there all those things are a reassembling of a God — that a God had to use itself up to create those things.

        That sort of absorbs the answer to the second question, too. If the explanation involves intent, the being with the intent is worthy of the title “God”. If the explanation doesn’t involve intent (like some area of physics) then we won’t call it God. But, the details of the explanation do matter.

        If it turns out we live in an intentionally created simulation, we will not call the programmers “God”, as they also belong to their own physical reality. In some sort of ‘simulation in a simulation in a simulation’ reality (hypothetical) it is the parent reality whose creation (with or without intention) matters.

        Are you happy with these parameters? If you are, feel free to start demonstrating the possibility/plausibility of God.

      14. You have agreed to the definition with several caveats. I assume you acknowledge this definition is not impossible.

        If it turns out we live in an intentionally created simulation, it is irrational to think the programmers existed within the code before they wrote it.

        If you consider irrational and rational explanations equally plausible, I will concede that I have no compelling arguments.

      15. I make no judgements about whether God, as defined above, is possible or not.
        It had not been demonstrated to be possible, but neither has it been demonstrated to be impossible.

        I understand the reference you are making about the code. It may become relevant to discuss the distinction between the universe as we know it and something analogous to Lawrence Krauss’ ‘Nothing’ or some other ‘pre-universal state’. But, given the way you defined the universe, it shouldn’t come up.

        I’m cautious about how you are already trying to set up any objection I have as ‘irrational’. There are no rational answers at the scale of question you’re ultimately looking at. I care about good explanations (defined as ‘well defined models that can both account for existing data and make specific predictions that can be explored).

        But the floor is still yours.

      16. You really didn’t get the message in this post, did you?
        Possibility is something that should be demonstrated, not assumed.
        By agreeing a definition, all we have done is define the claim. Demonstrating the claim is still all ahead of you.

      17. Actually, I did get the message of the post. In fact, this was what I guessed you would do. I would make a claim and you would say, “Impossible”.
        Turns out, I was wrong. You’re just going to shrug and tell me to keep talking.

      18. I really don’t know what sort of walk in the park you expected. I imagine you wanted the conversation to go like this:

        You: God is the explanation for the existence of the universe.
        Me: Okay. I accept this without reservation or question. Go for it.
        You: The universe exists. That needs an explanation. Therefore, by definition, God. Ta da.
        Me: Oh dang, I guess I’m a Christian now. Better go repent.

        Because, I can see why you’re frustrated if you think it was going to be a cake walk like that. But — dagnabbit — I have those pesky ‘question’ thingies. And it stops you having precisely the conversation you want, therefore ceases all conversation…

        Am I about hitting the nail on the head?

        Or is there some extent to which you actually realise there are slightly harder intellectual foundations to lay?

      19. LOL!
        When I ask questions, it’s an avoidance tactic.
        When you ask questions, it’s to advance the dialogue.
        Well played, Allallt! You have the makings of a fine Dear Leader.

      20. Tell you what, when I’m a leader, I will have actual points, policies and ideas to advance and that will put the burden of proof on me and I will actually address that then.

        In the mean time, while you’re the one with an idea or conclusion to advance, the burden of proof is on you.

        I mean, what possible use is my definition of God when I’m the one unconvinced there is one? It is only your definition that can advance the conversation.

        I almost feel like apologising for taking your idea seriously enough to actually challenge.

      21. Ah yes, the ultimate avoidance tactic of taking one’s bat and ball and going home… all the while blaming that taking your claims seriously enough to address them is precisely why atheists rarely move the conversation beyond “Possible vs. Impossible”.

  3. Great to have you back, Allallt!

    I have often pointed out the false equivalency between not impossible and probable/likely that the agnostic pretends exists in trying to establish a middle ground between believing and not believing in some god or gods, as if not knowing defends the goddy claim from justifying true belief beyond the not impossible. There’s something inherently dishonest here. It drives me nuts that this evasive tactic about belief used by the self-proclaimed agnostic is widely viewed as reasonable by those who wish to accommodate those who will not proclaim their belief state but hold it in abeyance while busy deriding those who admit an honest position as somehow less tolerant and more militant. The agnostic seems blissfully unaware that saying, “I don’t know what I believe but, gosh, I’m so much better a person for not taking a position at all than all of you lesser people who do” is really an admission of either not being honest or not daring to think at all. At the same time, the agnostic seems determined not to grasp that one’s current belief position is relative to the probability/likelihood of the belief claim being true. This is often where the same false equivalency is brought back into handy dandy use by the agnostic as the theist arguing that because we cannot know for certain some wildly improbable/unlikely claim is not true, therefore any belief claim might be true and we couldn’t possibly decide between them so we’d best go with a P=.5 position!

    This is an intentional and dishonest shifting of the fulcrum point of probability of some goddy belief claim from the middle (half way between zero as not true and one as true) to the not true end point of the probability spectrum and then claiming only this is the ‘balanced’ point where non belief is justified, that all other points on the belief spectrum belong to the agnostic regardless of probability and likelihood of the belief claim!

    Only in religion is this idiotic model of not willing to admit belief is granted any semblance of intellectual respect. Imagine trying to go through your day being met by the same reasoning, where the bridge builder doesn’t know if it’s safe to bear the weight so we’d best not cross it, the doctor doesn’t know if medication might work so better not take any, that the farmer doesn’t know if the food is any good so you’d better not eat any, that we just don’t know if the water is safe to drink so we’d better not drink any, and so on. Our justified agnosticism using this idiotic scale that demands either certainty or inaction would first paralyze us into inaction and inevitably result in our untimely deaths if followed with the same gusto so many faitheists want to insist is the only tolerant and reasonable way to think of religious claims. That’s why probability and likelihood matter when it comes to belief claims, and these do not belong only to the agnostic but remain a burden to the believer and an ally to the atheist. The agnostic just tries to avoid them altogether.

    1. I had never thought about this with respect to agnostics.
      I’m agnostic, in the technical sense that I don’t claim to know a God doesn’t exist. But I’m not so painfully (or smugly) unsure that I find the term agnostic useful in describing myself.
      The ‘I don’t know for sure and you can’t either’ position is something I see from theists more often, but that’s probably because that’s who I engage. The line of argument is normally ‘neither of us know there is no God; I have a string of words that I think makes God necessary; therefore I know there is a God’. It’s what Branyan is doing in the comments at the moment.
      But the agnostics giving cover to unsupportable superstition while also trying to allude to intellectual respectability is frustrating to read. And it does away with both: the reasons they don’t proclaim faith often dismantle the religious arguments, and the reason they don’t accept it is reasonable to reject faith is often intellectually vacuous.
      (And then there’s the Neil deGrasse Tyson ‘agnostic’ who asserts that, to him, atheism is about going out and debating religious people, and so that’s not him… Also irritating.)

  4. As you point out, most of these “arguments” smuggle a god into their premises and Voila! one pops out as the conclusion! Can there be a maximally great anything? Or is the concept solely an imaginary thing? Well, think about all of the stars in the sky. There are hundreds of billions of galaxies, each with hundreds of billions of stars. We could, theoretically, arrange them all by size (say their diameter). At one end of the sequence, we would have the smallest star in the universe and at the other end the largest one. As with many such things, there is a lower limit on diameter of a star. You can’t have a star with a diameter of zero. But is there an upper limit? Could there not be a star a little bigger than the biggest we found? More than likely this is possible.

    Similarly with temperatures, there is a lower bound, but not an upper bound. So, the idea of a maximally great anything is bogus. (You say your god is maximally great and I say, no He has a Mom!) If you imagine a star bigger than the known universe, I can imagine a larger universe. (Neener neener.) So, such arguments are bogus and they are just sneaking their god into the premises (all assumed to be true, otherwise they would be premises) so they can pull it out of their hat (or ass) at the end.

  5. “You say your god is maximally great and I say, no He has a Mom!”
    This is a misunderstanding of the argument.

    A maximally great being is different from a “maximally big star”. The being is hypothetically the source of every lesser entity (like stars and temperatures). This great “being” cannot be a material entity within the universe like a mountain, a star or a galaxy because that makes the universe exist without an explanation.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s